My 'new' McIntosh MI-3

AdamAnt316

Collector of heavy things
Subscriber
Hello everyone! I was looking around at one of the local hamfests when what should I spot sitting on a seller's table but a McIntosh MI-3! It was sitting on top of a Heathkit AJ-15 tuner; the seller was willing to sell it separately for $200. According to the seller, it has HV and/or CRT issues. He told me it had worked fine with the Heathkit tuner until one day, when the MI-3's fuse blew.

He was under the impression that the HV selenium rectifier may have been to blame, took it out-of-circuit, and replaced it using three silicon diodes in series (1N4007s, I'm guessing) along with a resistor in an attempt to make up for the difference voltage drop. This had apparently cured the fuse blowing (not sure how many times it happened), but the CRT still wouldn't might up, which made him think that the CRT may have been blown. He picked up a replacement 3RP1 from Sphere Research, which was included.

Using a cheapo DMM, I checked the filament of the CRT, which turns out to be OK. I also ohmed out the HV taps on the power transformer: The main HV tap (1200V?) measured around 1,100 ohms, while the blue wire for the center tap (400V?) measured around 80 ohms to the red/yellow wire. Does this seem out-of-place? Is there a known source for a replacement, or might there be a company out there which would be willing to rewind a transformer like this? Anyway, here are some photos:

Front of the unit:
mi3_front.jpg

The front panel looks to be in good shape overall:
mi3_panel.jpg

A view of the top of the chassis, seems pretty clean:
mi3_top.jpg

Underside of the chassis, again looks pretty clean:
mi3_bottom.jpg

Closeup of the chassis; note the modification in the HV section:
mi3_chassis.jpg

Close-up of the modification:
mi3_diode1.jpg

Another angle of the mod:
mi3_diode2.jpg


Other than the diode mod, it looks pretty good inside. I visited a local electronics supply store, and they had a rather large NTE-517 diode there meant for microwave ovens. It's rated for 15KV@550mA, so it should be more than enough to replace the old selenium diode. Might the old diode going bad have caused the issue, or is it more likely to be the power transformer? Any suggestions on how to go forward? Thanks in advance!
-Adam
 
I picked up one of these last year and was told it didn't work. In mine, the 6U8 tubes were bad (not low testing, NO testing.. dead dead) I replaced them and did as you did and replaced the selenium rectifier with silicone and it now works perfectly.
Not sure if these are rough on these particular tubes, but I found it weird they were all shot like that. Good luck.

If you want, I will take a shot of under the hood on mine.
 
Last edited:
Hello everyone! I was looking around at one of the local hamfests when what should I spot sitting on a seller's table but a McIntosh MI-3! It was sitting on top of a Heathkit AJ-15 tuner; the seller was willing to sell it separately for $200. According to the seller, it has HV and/or CRT issues. He told me it had worked fine with the Heathkit tuner until one day, when the MI-3's fuse blew.

He was under the impression that the HV selenium rectifier may have been to blame, took it out-of-circuit, and replaced it using three silicon diodes in series (1N4007s, I'm guessing) along with a resistor in an attempt to make up for the difference voltage drop. This had apparently cured the fuse blowing (not sure how many times it happened), but the CRT still wouldn't might up, which made him think that the CRT may have been blown. He picked up a replacement 3RP1 from Sphere Research, which was included.

Using a cheapo DMM, I checked the filament of the CRT, which turns out to be OK. I also ohmed out the HV taps on the power transformer: The main HV tap (1200V?) measured around 1,100 ohms, while the blue wire for the center tap (400V?) measured around 80 ohms to the red/yellow wire. Does this seem out-of-place? Is there a known source for a replacement, or might there be a company out there which would be willing to rewind a transformer like this? Anyway, here are some photos:

Front of the unit:
mi3_front.jpg

The front panel looks to be in good shape overall:
mi3_panel.jpg

A view of the top of the chassis, seems pretty clean:
mi3_top.jpg

Underside of the chassis, again looks pretty clean:
mi3_bottom.jpg

Closeup of the chassis; note the modification in the HV section:
mi3_chassis.jpg

Close-up of the modification:
mi3_diode1.jpg

Another angle of the mod:
mi3_diode2.jpg


Other than the diode mod, it looks pretty good inside. I visited a local electronics supply store, and they had a rather large NTE-517 diode there meant for microwave ovens. It's rated for 15KV@550mA, so it should be more than enough to replace the old selenium diode. Might the old diode going bad have caused the issue, or is it more likely to be the power transformer? Any suggestions on how to go forward? Thanks in advance!
-Adam

Hi Adam...

Typically speaking failure of the power transformer is rare. While the 6U8's are relatively reliable it should be remembered that proper ventilation is also important. In your post, you indicated that your used 1N-4007 diodes. While these will work, their current rating is marginal. In our upgrades we use GI-828 devices. Also important is that the voltage drop across the selenium's is greater than the drop across the silicon devices. With the change of diodes make sure that you also install an additional limiting resistor between the positive output of the diodes to the first stage filtering. Adjust this value to the original voltage plus or minus 5%. In addition, you should check the value of the power supply filters. Generally speaking, with the restoration, these are replaced. As far as the CRT issue, you should check the Horizontal and Vertical position devices for condition. Hope this is of some help.

Mark
McIntosh Factory Restoration & Warranty Service Facility
Phone:(952) 894-5580
Fax: (952) 456-4937
eMail: McIntosh@AbSoundLabs.Com
 
Thanks for the replies.
I picked up one of these last year and was told it didn't work. In mine, the 6U8 tubes were bad (not low testing, NO testing.. dead dead) I replaced them and did as you did and replaced the selenium rectifier with silicone and it now works perfectly.
Not sure if these are rough on these particular tubes, but I found it weird they were all shot like that. Good luck.

If you want, I will take a shot of under the hood on mine.

Is yours an MI-2 or an MI-3? IIRC, the MI-2 used 6U8s, while the MI-3 used 6EA8s. I have yet to check them, since I don't think they're the cause for the fuse blowing, but I'll do so anyway. All of the tubes in my MI-3 look to be originals, with yellow McIntosh markings.

Does yours have a tube rectifier for the HV, or a solid-state one like mine? Pictures would be nice for comparison purposes, at the least, particularly where the underside is concerned. I'd also appreciate it if you could take a resistance reading across the red/red-yellow and blue/red-yellow wires of the transformer in yours, with the power turned off and the filter capacitors drained to prevent inaccurate readings (not to mention protecting your meter from getting blown up).

Hi Adam...

Typically speaking failure of the power transformer is rare. While the 6U8's are relatively reliable it should be remembered that proper ventilation is also important. In your post, you indicated that your used 1N-4007 diodes. While these will work, their current rating is marginal. In our upgrades we use GI-828 devices. Also important is that the voltage drop across the selenium's is greater than the drop across the silicon devices. With the change of diodes make sure that you also install an additional limiting resistor between the positive output of the diodes to the first stage filtering. Adjust this value to the original voltage plus or minus 5%. In addition, you should check the value of the power supply filters. Generally speaking, with the restoration, these are replaced. As far as the CRT issue, you should check the Horizontal and Vertical position devices for condition. Hope this is of some help.
That's good to know, thanks. My MI-3 is supposed to use 6EA8s, according to the service manual; I don't know off-hand why McIntosh made the switch from 6U8s. What is the difference between the two tube types? The Tube Store has them listed together, while RadioMuseum says the 6EA8 is equivalent to the 6GH8, but says nothing about the 6U8. I'll have to open my MI-3 back up to see exactly what they used in it.

I've heard that earlier versions (particularly the MI-2) had issues with transformer failure due to the 6W4 rectifier tube developing issues. The solution was to replace it with a solid-state version, which mine seems to have had from the factory since there's no sign of there having been a tube socket for it. Whoever subbed in the silicon diodes installed a resistor of some sort in series with them, though I dunno if it was the proper value (they were going by a guide for replacing selenium rectifiers with silicon diodes in old radios). The data sheet for the NTE-517 diode I bought lists the forward voltage drop (VFM) as 14V, though this page lists it as 610mV. Not sure which one is correct, though 14V seems akin to the sort of voltage drop which would typically be seen when using a selenium diode. Am I correct on this?

The high resistance reading I'm getting across the HV winding seems a bit odd, though it's with everything wired in place, so maybe that could throw things off. Is there a better way to check whether or not the transformer is operating correctly without resorting to a HV probe? I was planning to connect my HP multimeter (rated for 1000V max) across the red and blue wires from the transformer, and see what AC voltage reading I get as I bring the unit up on a variac. Does this seem like a good idea, or would there be a better way to go about this? Again, thanks!
-Adam
 
I have the MI-3. It had 6u8's that were dead and I replaced with 6U8. The silkscreen does say 6ea8 though, like you said. (I believe it's pretty much the same tube)

I will take a look under the hood when I get home..

IMG_9720.JPG


IMG_9721.JPG
 
That's good to know, thanks. My MI-3 is supposed to use 6EA8s, according to the service manual; I don't know off-hand why McIntosh made the switch from 6U8s. What is the difference between the two tube types? The Tube Store has them listed together, while RadioMuseum says the 6EA8 is equivalent to the 6GH8, but says nothing about the 6U8. I'll have to open my MI-3 back up to see exactly what they used in it.

I think the 6ea8, 6gh8, 6U8 and a few others are all inter changeable. Like I said, I put 6U8's into that location without any issues. My RCA tube data book list them as all replacements for one another. I'm by no means an expert on the subject. Someone will stumble along that can tell you more..
 
Again, thanks. I popped the 'hood' on my MI-3, and all five of the tubes are still 6EA8s, all McIntosh-branded. I'll have to test them with my Heathkit TC-2 tube tester to see if any of them have gone bad. Guessing the settings for 6U8 would work; will have to see if this type is listed on any of the charts which came with it.

Anyway, I opened the bottom, and did some readings on the HV diodes. The old selenium HV diode appears to be shorted. I removed it, along with the string of 1N4007s (which were paired with a 10K resistor, brown-black-red), and installed the NTE517 as the old EDAL rectifier would've been (striped end towards the red wire from the transformer). I connected the leads of my HP DMM to the red and blue wires coming from the power transformer, set the meter to AC volts, plugged the MI-3 into a Variac, and started slowly turning it up. At around 12V in, I was reading 80V on the meter, meaning that the HV winding appears to be OK! :)

However, that's when things started to go pear-shaped. I turned the Variac knob back down, reconnected the clip leads across the AC IN, and slowly started cranking things up. At around 20-30V, I started seeing light from the 0B2 regulator tubes, with one of them lighting up quite brightly and flickering. I also started hearing some crackling noises from somewhere. Unfortunately, I soon noticed some smoke rising from the area of the intensity control! :eek: I quickly powered things down, and let the smoke dissipate. Obviously, something's still quite wrong with this unit, likely off of the -1300V line (which is what the intensity control is connected to). Any ideas what I might want to check next? Again, thanks in advance!
-Adam
 
Thanks for the replies.


Is yours an MI-2 or an MI-3? IIRC, the MI-2 used 6U8s, while the MI-3 used 6EA8s. I have yet to check them, since I don't think they're the cause for the fuse blowing, but I'll do so anyway. All of the tubes in my MI-3 look to be originals, with yellow McIntosh markings.

Does yours have a tube rectifier for the HV, or a solid-state one like mine? Pictures would be nice for comparison purposes, at the least, particularly where the underside is concerned. I'd also appreciate it if you could take a resistance reading across the red/red-yellow and blue/red-yellow wires of the transformer in yours, with the power turned off and the filter capacitors drained to prevent inaccurate readings (not to mention protecting your meter from getting blown up).


That's good to know, thanks. My MI-3 is supposed to use 6EA8s, according to the service manual; I don't know off-hand why McIntosh made the switch from 6U8s. What is the difference between the two tube types? The Tube Store has them listed together, while RadioMuseum says the 6EA8 is equivalent to the 6GH8, but says nothing about the 6U8. I'll have to open my MI-3 back up to see exactly what they used in it.

I've heard that earlier versions (particularly the MI-2) had issues with transformer failure due to the 6W4 rectifier tube developing issues. The solution was to replace it with a solid-state version, which mine seems to have had from the factory since there's no sign of there having been a tube socket for it. Whoever subbed in the silicon diodes installed a resistor of some sort in series with them, though I dunno if it was the proper value (they were going by a guide for replacing selenium rectifiers with silicon diodes in old radios). The data sheet for the NTE-517 diode I bought lists the forward voltage drop (VFM) as 14V, though this page lists it as 610mV. Not sure which one is correct, though 14V seems akin to the sort of voltage drop which would typically be seen when using a selenium diode. Am I correct on this?

The high resistance reading I'm getting across the HV winding seems a bit odd, though it's with everything wired in place, so maybe that could throw things off. Is there a better way to check whether or not the transformer is operating correctly without resorting to a HV probe? I was planning to connect my HP multimeter (rated for 1000V max) across the red and blue wires from the transformer, and see what AC voltage reading I get as I bring the unit up on a variac. Does this seem like a good idea, or would there be a better way to go about this? Again, thanks!
-Adam
The MI-1(Limited production and not marketed) and the MI-2and early versions of the MI-3 did have issues caused by the implantation of the rectifier tube. That tube, in further production was modified with a diode assembly. substitution. This issue did not occur in the later production MI-3's and MI-4's instruments. Details for this modification can be found in McIntosh Service Bulletin #110.

I am not sure where the 14 volt drop came from. The drop across a silicon diode is 0.6 volt and on a germanium diode is 2.25 volts.The original diodes have a 3000 volt PIV @ 3.0 ma.

Regarding the 6U8 question. The 6U8/6EA8/6GH8 tubes are very similar in characteristics. Since these tubes are not in any audible circuitry sonics isn't really of any issue. I recall that McIntosh used both 6U8's and 6EA8's, but my mind might be fogged as I am really getting old, after being in this industry for 51 years.

As far as the issue with the popping and smoke, this can be caused by several failures. As such I can't comment there.

Hope this is of some help.

Mark
McIntosh Authorized Restoration & Warranty Service Facility
Phone:(952) 894-5580
Fax: (952) 456-4937
eMail: McIntosh@AbSoundLabs.Com
 
Again, thanks. I popped the 'hood' on my MI-3, and all five of the tubes are still 6EA8s, all McIntosh-branded. I'll have to test them with my Heathkit TC-2 tube tester to see if any of them have gone bad. Guessing the settings for 6U8 would work; will have to see if this type is listed on any of the charts which came with it.

Anyway, I opened the bottom, and did some readings on the HV diodes. The old selenium HV diode appears to be shorted. I removed it, along with the string of 1N4007s (which were paired with a 10K resistor, brown-black-red), and installed the NTE517 as the old EDAL rectifier would've been (striped end towards the red wire from the transformer). I connected the leads of my HP DMM to the red and blue wires coming from the power transformer, set the meter to AC volts, plugged the MI-3 into a Variac, and started slowly turning it up. At around 12V in, I was reading 80V on the meter, meaning that the HV winding appears to be OK! :)

However, that's when things started to go pear-shaped. I turned the Variac knob back down, reconnected the clip leads across the AC IN, and slowly started cranking things up. At around 20-30V, I started seeing light from the 0B2 regulator tubes, with one of them lighting up quite brightly and flickering. I also started hearing some crackling noises from somewhere. Unfortunately, I soon noticed some smoke rising from the area of the intensity control! :eek: I quickly powered things down, and let the smoke dissipate. Obviously, something's still quite wrong with this unit, likely off of the -1300V line (which is what the intensity control is connected to). Any ideas what I might want to check next? Again, thanks in advance!
-Adam

Coincidentally, I have a TC-1 and a TC-2 tube tester, and they are. If you need the settings let me know.
 
The MI-1(Limited production and not marketed) and the MI-2and early versions of the MI-3 did have issues caused by the implantation of the rectifier tube. That tube, in further production was modified with a diode assembly. substitution. This issue did not occur in the later production MI-3's and MI-4's instruments. Details for this modification can be found in McIntosh Service Bulletin #110.

I am not sure where the 14 volt drop came from. The drop across a silicon diode is 0.6 volt and on a germanium diode is 2.25 volts.The original diodes have a 3000 volt PIV @ 3.0 ma.

Regarding the 6U8 question. The 6U8/6EA8/6GH8 tubes are very similar in characteristics. Since these tubes are not in any audible circuitry sonics isn't really of any issue. I recall that McIntosh used both 6U8's and 6EA8's, but my mind might be fogged as I am really getting old, after being in this industry for 51 years.

As far as the issue with the popping and smoke, this can be caused by several failures. As such I can't comment there.

Hope this is of some help.

Mark
Again, thanks for the reply. I've seen that service bulletin, and there are a few posts on various forums around the 'net about people looking for new power transformers for their MI-3s. That got me worrying about the condition of the transformer in this MI-3, but as I stated earlier, that appears to be an unfounded concern, fortunately.

Good to know on the tubes. You are correct; the MI-2 was originally spec'd for 6U8s, while McIntosh decided to switch to 6EA8s for the MI-3 for some reason. Mine does have 6EA8s installed; haven't yet checked to see whether or not they're any good, but they're unlikely to be the cause of the issue I'm seeing at the moment, AFAIK.

I got my 14VFM figure from the datasheet for the NTE517 diode I'm using to replace the EDAL G4R48H diode originally used in this device. The original selenium diode is reading a direct short; whether or not this is due to whatever's shorted ahead of it I'm not entirely sure. As for what might be shorted, there doesn't seem to be a whole lot ahead of the intensity control on the schematic; either the CRT has a heater-to-cathode short, or something is touching the chassis which shouldn't be. I might have to dig out my old CRT checker, and see if it's up to the task..........
-Adam
 
FURTHER UPDATE:
It appears I was incorrect in stating that the power transformer in this MI-3 seems to be good. :( After talking with some folks who know more about this stuff than I do, I unsoldered the wire which connects the heater/cathode of the CRT to one end of the intensity pot, unplugged the CRT socket, and tried bringing up the voltage again. This time around, the intensity pot didn't smoke, and I was getting HV from the power supply as normal. Apparently, this indicates that the filament winding for the CRT has internally shorted to the HV winding, which apparently was a common problem in early television sets.

Since I'm unlikely to find a replacement transformer for this unit, it has been suggested to me to install a separate 6.3V filament transformer (one which provides some manner of isolation against the high voltage which part of the CRT's filament is at the potential of, like a split-bobbin type) to bypass the shorted winding and light the CRT. Apparently, this was a common workaround fix in the early days of television, since it was faster to do than ordering a replacement power transformer for the set, which would've meant that the TV repairman's customer might miss the latest episode of "I Love Lucy".......... :rolleyes:
-Adam
 
Minor update: I've ordered a Hammond 166G6B transformer, which is a split-bobbin design rated for the same current as the CRT's filament draw. According to the datasheet, it has been hi-pot tested to 2,000V, so it should be able to survive the -1,300V riding on one side of the filament. Not quite sure where I'm going to mount it, since I'd rather not drill holes in the chassis, but it should do the trick.
-Adam
 
Another minor update: the Hammond transformer has arrived in the mail, and I will be test-fitting it soon. I've also acquired a full set of 6EA8 tubes for this MI-3, plus a replacement for the 8µF@475V filter capacitor (actually rated for 10µF@600V) which should do the trick. I've also acquired a B&K 466 CRT tester which hopefully can be used to test the existing 3RP1; hopefully, this incident hasn't damaged it internally. With any luck, I'll get some sort of trace on the screen before too long. :)
-Adam
 
Hi Adam...

Be vary careful if you use any type of CRT tester or rejuvenator. Many were not designed for any type of CRT other than a TV CRT. Specifically, do not try to do this action with the Sencore line of instruments. If you need the original, I have seen them, from time to time, on eBay.

Mark
McIntosh Factory Service Center
.
 
Hi Adam...

Be vary careful if you use any type of CRT tester or rejuvenator. Many were not designed for any type of CRT other than a TV CRT. Specifically, do not try to do this action with the Sencore line of instruments. If you need the original, I have seen them, from time to time, on eBay.

Mark
McIntosh Factory Service Center
.
Thanks for the reply. Most of the info I've seen online suggests that the B&K 466 and its ilk are not meant for testing the emissions of electrostatic deflection CRTs like the 3RP1 in the MI-3. My main hope at this point is to use the 466 to check whether or not the high-voltage issues may have damaged the insulation between the heater and cathode within the CRT itself. Granted, the cathode is tied to one side of the heater within the MI-3's CRT socket, so it may not matter too much, but I'd still like to be able to check it if I don't get a solid trace once I get it working. What are you referring to by 'the original'?
-Adam
 
Hi Adam...

Sorry for the slip. I was referring to the 3RP1 CRT.

Mark
McIntosh Factory Service Center
Understood. This MI-3 came with a NIB Chinese-made 3RP1 which, as far as I know, has yet to be taken out of its original box. I'm not entirely sure about the condition of the existing CRT; as far as I know, the filament has continuity (the seller thought the HV issue might've blown it) and doesn't show any obvious shorts. The B&K 466 tester has the ability to check for H-K shorts, among other things, and I'm hoping it can be used to test the existing 3RP1 in this capacity.
-Adam
 
Yet another update: After a fair amount of preparation, I went ahead and installed the above-mentioned Hammond filament transformer under the hood. Looking to avoid drilling new holes or whatnot, I used a series of metal brackets and plastic standoffs to mount the new transformer between the standard power transformer and the square can cap using the latter's mounting screws. I ran the wires through a couple of the vent holes in the chassis, unsoldered the old wires from the CRT winding, and soldered in the new transformer. I slowly brought up voltage to the unit using my variac, and not only did it not smoke, I got a nice green dot on the CRT! :banana: I used a Y-cable to connect an old iPod to the audio in jacks, and got a nice squiggly waveform on the screen. I have yet to pronounce the unit completely done (probably going to replace the HV filter cap first, and the intensity control has some dead spots), but it's getting pretty close! :D
-Adam
 
After my previous post, I went ahead and replaced the aforementioned filter capacitor with a modern 10µF 600V cap, retaining the old Sprague for originality purposes. Not entirely sure that the unit is good to go just yet, but I'll probably be installing it in my McStack soon! :biggrin: Here are some new pictures of the MI-3:

Top of the chassis, with the new CRT filament transformer installed:
mi3_top2.jpg


Better view of the new transformer. The brackets I used are thinner than I'd like, so I put reinforcing clips over the top of them while I look for better ones. Not quite Mc factory work, but completely reversible:
mi3_newtrans1.jpg


New shot of the chassis underside. Looks mostly the same, apart from the wires from the new transformer, and the new diode and capacitor in the HV supply:
mi3_chassis2.jpg


Close-up of the new NTE517 diode and filter capacitor:
mi3_newdiode.jpg


And the moment we've all been waiting for, the MI-3 in action, attempting to parse the waveform of Donald Fagen - I.G.Y. via my iPod: :music:
mi3_running.jpg
 
Back
Top Bottom